ON THE INFLUENCE OF COSMIC NOISE ON THE HOURLY NUMBER OF DETECTED METEORS

L. P. SERAFINOVICH

(Presented by the scientific seminar of the Radio Engineering Faculty)

Recently, a number of articles have appeared in the foreign literature devoted to discussion of the question of the influence of variations in the intensity of cosmic noise on the results of radar observations of meteors.

Moreover, these articles contain contradictory points of view on this question. Thus, for example, the Australian investigators Elliott and Fraser [1] carried out an experiment at a frequency of 69 MHz and found that cosmic noise affects the observed hourly number of meteors much more strongly than the receiver’s internal noise. They point to the necessity of taking into account the effective temperature of the antenna when comparing hourly numbers of reflections, even if these numbers were obtained on the same equipment. The same opinion is held by Shine and Kerr [2], who worked at a frequency of 18 MHz and introduced corrections for cosmic noise into the observed hourly numbers of reflections.

However, Lovell [3], while confirming the possibility of the influence of cosmic noise at a frequency of 18 MHz, indicates that at a frequency of 70 MHz the influence of the same noise on the sensitivity of the apparatus, and consequently on the observed hourly numbers of reflections, is small and can easily be taken into account in the analysis.

Hawkins [4], to whose work Lovell refers, likewise indicates that at a frequency of 70 MHz there were no noticeable changes in the sensitivity of the apparatus caused by cosmic noise. This, as is evident, completely contradicts the experimental data of Elliott and Fraser.

In connection with the above, it seems to us advisable once again to consider the question of the influence of variations in the intensity of cosmic noise on the hourly numbers of reflections. For this purpose we shall use the data given in Hawkins’s article [4], namely: the mean monthly daily distribution of hourly numbers of reflections for sporadic meteors for the period from October 1950 to September 1951 (Fig. 1). These distributions were obtained with the apparatus of the Jodrell Bank station (longitude \(2^\circ 18' W\), latitude \(53^\circ 14' N\)). The observations were carried out using two antennas simultaneously: the first antenna was directed to the southwest (azimuth \(242^\circ\), measured from north clockwise); the second—to the northwest (azimuth \(292^\circ\)). Both an-

directed and give identical beams with elevation angle \(h \simeq 8^\circ 30'\) and a beam width of \(\sim 10^\circ\) at half power in the vertical and horizontal planes.

Fig. 1. Mean monthly diurnal curves of the distribution of hourly numbers of reflections for 1950–1951.

The apparatus operated at a frequency of 70 Mc/s with a minimum detectable signal

\[ P_{\min}=7\cdot 10^{-14}\ \mathrm{W}\ [5]. \]

Using the data given, we determine the variation of the temperature of each antenna during the day. To do this, we find the paths of the rays

for both antennas over the celestial sphere and, using the nearest isophotes of cosmic radio emission at a frequency of 64 MHz [6], compute the antenna temperatures for each hour of the day.

The data obtained are shown in Fig. 1.

The use of isophotes at a frequency of 64 MHz introduces a small quantitative error, but does not change the picture qualitatively.

We next correct the diurnal distribution of hourly numbers of echoes, given by Hawkins, for the obtained variations of antenna temperatures over the course of the day.

For this we shall use the formula [7]

\[ N_o = N \cdot Q, \]

where \(N_o\) is the true number of meteor echoes;

\(N\) is the number of meteor echoes recorded by the apparatus;

\[ Q \simeq \left[\frac{T_a(t)}{T_{a \min}}\right]^{\frac{s-1}{2}} \]

is a correction allowing for the change in noise intensity over the course of the day;

\(s\) is a constant characterizing the law of distribution of meteor bodies by mass. For sporadic meteors \(s \simeq 2\);

\(T_a\) is the antenna temperature.

The corrections are introduced relative to the mean minimum noise level \(T_{a \min}\) during the day. Thus, for the 1st antenna we take \(T_{a \min} \simeq 2500^\circ\mathrm{K}\), and for the 2nd antenna \(T_{a \min} \simeq 2700^\circ\mathrm{K}\). Taking the minimum level of cosmic noise during the day as the reference level makes it possible to obtain a diurnal distribution of the number of echoes that is closer to the real one. The corrected results are shown in Fig. 1.

Thus, Fig. 1 gives, for both antennas, the diurnal distribution of hourly numbers of echoes obtained by Hawkins, and the diurnal distribution of hourly numbers of echoes computed from the formula given above, as well as the variations of antenna temperatures. Let us carry out an analysis.

When viewing the graphs for all months, one can immediately note the fact that the maxima of noise fall on the depressions in the hourly numbers of echoes. Consequently, the latter can be explained by an increase in the intensity of cosmic noise during these hours. This is especially clearly noticeable in May and June for the 1st antenna. Lovell, in his note [3], points to the absence in May–June of a peak at 6:00 in the morning. However, the corrected distribution for the 1st antenna shows that this peak in May–June actually exists, but is masked by noise. For the 2nd antenna in May, a maximum is also observed at 6:00 in the morning.

In these same months there is a considerable predominance of daytime maxima in the number of echoes, especially for the 2nd antenna. The introduction of noise corrections shows that the daytime meteor activity already begins to appear in March and increases in April. The correction also makes it possible to obtain a smoother increase in the number of echoes for the 2nd antenna during the daytime hours from April to June.

An interesting phenomenon is observed in July. The graph of the number of echoes for the 2nd antenna shows a considerable predominance of the nighttime peak over the daytime one. For the 1st antenna this is not observed, although in all other months there is a clear correlation between the two antennas. The introduction of noise corrections shows that such

the distortion is obtained for the 1st antenna owing to an increase in the intensity of cosmic noise at this time.

In the corrected curves there is observed a distinct, as yet unexplained, nocturnal maximum in the period June–February, lying within ±4 hours relative to midnight, depending on the month. This maximum is absent from the curves given by Hawkins in August–September, where it is evidently suppressed by noise. If noise were absent, it would be unclear why such a strong July maximum suddenly disappears in August–September and then a weaker one appears again in October–February. In the corrected case, the decrease of the nocturnal maximum occurs gradually from June to February. This nocturnal maximum considerably exceeds the peak at 6:00 in the morning in July–September, merging with the latter. Therefore the usual morning maximum of reflections seems to disappear.

Fig. 2. Mean annual diurnal curves of the distribution of hourly numbers of reflections for 1950–1951.

The influence of cosmic noise can also be detected from the following phenomenon. If one follows the motion of the minimum in the number of reflections in January, February, and March, it may be noted that it is synchronous with the motion of the noise maximum. Thus, for the 1st antenna the minimum of reflections falls: in January at 15 h, in February at 13 h 30 min, in March at 11 h 30 min; for the 2nd antenna—in January at 18 h 30 min, in February at 16 h 30 min, in March at 15 h 00 min. The maxima of cosmic-noise intensity in these months fall at the same hours. When the latter are taken into account, the minima in the number shift toward 18 hours, which is observed in almost all the other months.

This is clearly visible from the curves for the 2nd antenna. The exceptions are, for the 1st antenna, the months of October and November, and for the 2nd antenna—December.

In these months the maxima of the intensity of cosmic noise occur 1.5–2.5 hours after 18 h 00 min, coinciding with the minima of the number of reflections. Introducing corrections for noise leads to the minima being shifted to 15 hours.

Thus, in the corrected curves the minima in the number of reflections are located, in all months, within the range 15–18 hours, which is entirely understandable.

From the analysis performed it is evident that, in the present case, cosmic noise has a substantial influence on the qualitative and quantitative diurnal distribution of the hourly numbers of reflections at a frequency of 70 MHz. The indication by Elliott and Fraser of the need to take the antenna temperature into account when comparing hourly numbers of reflections is confirmed. At the same time, Lovell’s assertion that the influence of cosmic noise at a frequency of 70 MHz is small is refuted.

In his note [3], Lovell also pointed out the erroneousness of the opinion that sporadic meteors have a uniform distribution, in which meteor activity has a maximum at 06 h 00 min and a minimum at 18 h 00 min local time. This is confirmed by the corrected curves, from which it is evident that daytime meteor streams (mentioned by Lovell) and nighttime meteors have an important influence on the diurnal distribution of meteors.

Let us next consider how cosmic noise affects the diurnal distribution of the hourly numbers of reflections taken over a year. In Fig. 2 such curves, obtained by Hawkins, are presented, together with curves corrected for noise.

As is evident from a comparison of the two types of curves, the change in the intensity of cosmic noise over the course of a day does not affect the qualitative picture of the diurnal distribution of reflections taken over a year. Only the hourly numbers of reflections change slightly.

In conclusion it should be noted that, since the increase in the intensity of cosmic noise occupies a relatively short interval of time during the day (\(\sim 3\) hours), it apparently was not noticed when measuring the minimum detectable signal. The indicated signal \(P_{\min}=7\cdot 10^{-14}\) W corresponds to \(T_a \approx 3000^\circ\mathrm{K}\), which is observed during the rest of the day, whereas at \(T_a \approx 13000^\circ\mathrm{K}\), \(P_{\min}=2\cdot 10^{-13}\) W.

Conclusions

1. The change in the intensity of cosmic noise over the course of a day has a substantial influence on the hourly number of reflections.

2. This change in the intensity of noise fundamentally changes the diurnal and mean monthly diurnal distribution of the hourly numbers of reflections, especially when the number of reflections is large.

3. This change does not substantially affect the diurnal distribution of the hourly numbers of reflections taken over a year.

4. With significant changes in the intensity of cosmic noise, the number changes only slightly (thus, for \(T_a\) changing by a factor of 5, \(N\) changes by only a factor of 2).

5. The change in the intensity of cosmic noise over the course of a day should be taken into account at all frequencies below 100 MHz, and corrections should be introduced into the diurnal distribution of the hourly numbers of reflections. This is especially important if the antenna beam passes through regions of the celestial sphere lying near the center of the Galaxy.

6. As the reference level when introducing corrections, one should take the minimum level of the intensity of cosmic noise during ...

days. This gives a distribution of the number of reflections from meteors over the course of a day that is closer to the real one.

LITERATURE

1. Ellyett C. D., Fraser G. J. Austr. J. Phys., v. 8, No. 2, p. 273—278, 1955.
2. Shain C. A., Kerr F. J., J. Atm. Terr. Phys., v. 6, p. 280, 1955.
3. Lovell A. C. B. J. Atm. Terr. Phys., v. 8, N 4/5, p. 293—294, 1956.
4. Hawkins G. S. Mon. Not. of Royal Astr. Soc., v. 116, No. 1, p. 92—104, 1956.
5. Aspinall A., Clegg J. A., Hawkins G. S., Phill. Mag., v. 42, p. 504, 1951.
6. Pawsey, Bracewell, Radio Astronomy. Oxford, 1955.
7. E. I. Fialko, L. P. Serafinovich. On the influence of the intensity of cosmic interference on the hourly number of meteors detected by radio means. Izvestiya TPI, vol. 105, p. 28, 1960.